ops-bottleneck-review
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill is entirely descriptive, consisting of Markdown checklists and YAML configuration. It contains no Python scripts, Node.js code, or shell commands.
- [INDIRECT_PROMPT_INJECTION]: The skill processes business workflow data which could theoretically contain instructions.
- Ingestion points: User-provided workflow and customer promise data in SKILL.md.
- Boundary markers: Absent.
- Capability inventory: No file-write, network, or command execution capabilities found in any file.
- Sanitization: Absent.
- [SAFE]: No indicators of obfuscation, exfiltration, or persistence were found. The skill operates within a restricted scope focused on business analysis.
Audit Metadata